kvanttt / basencoding Goto Github PK

I'm using this library as part of a project of mine, and I've found in some cases there would be a need to validate an encoded string, that is, make sure that all of the characters are valid and that the formatting is correct.

Would this be possible to add?

See here:
https://en.bitcoin.it/wiki/Base58Check_encoding
https://github.com/bitcoin/bitcoin/blob/master/src/base58.h

There is a huuuge bug here:

For example, if you enter "ap\n" ("ap" and NEW LINE) here:

http://kvanttt.github.io/BaseNcoding/

using Base=4096 and MaxBitCount=12 it will encode into:

ٚ+

which decodes back into "az" instead of its original value "ap\n" ("ap" and NEW LINE).

This bug is not related to web site character encoding nor new line differences between OSs (\r, \n vs \r\n) but it is easily reproducable with unit tests.

Hello,
First of all, thanks for this nice Library, It has really been helpful.
I and my colleague @ron4fun found a bug in the encoding and decoding of certain values to Base32.
A test example is the Name "Ronald" without the quotes, verifying with other encoders gives "KJXW4YLMMQ======" but yours gives "KJXW4YLMMQ====" which is wrong.
I have fixed the issue in my Pascal Port but just decided to inform you about the bug.
I could provide a pull request if you want.
Regards.

I have a 169 bytes image. Lets encode it.

Just run it in the browser console (on https://kvanttt.github.io/BaseNcoding/):

let dataUrl = `data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAACAAAAAgCAIAAAD8GO2jAAAAAXNSR0IArs4c6QAAAARnQU1BAACxjwv8YQUAAAAJcEhZcwAADsMAAA7DAcdvqGQAAAA+SURBVEhL7dIxCgAwCMVQ739pu/yh0JYgXfNWlSxWv9UhgwkDyAAygAygXEo/8k032dhkMGEAGUAGkAHQvQC3veR+Ql0lAQAAAABJRU5ErkJggg==`;
let base64 = dataUrl.slice("data:image/png;base64,".length);
let binaryString = atob(base64);

var alphabet = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/";
var baseN = new BaseN(alphabet, 64);
baseN.encodeString(binaryString);

The result is:

wolQTkcNChoKAAAADUlIRFIAAAAgAAAAIAgCAAAAw7wYw63CowAAAAFzUkdCAMKuw44cw6kAAAAEZ0FNQQAAwrHCjwvDvGEFAAAACXBIWXMAAA7DgwAADsODAcOHb8KoZAAAAD5JREFUSEvDrcOSMQoAMAjDhVDDr39pwrvDvMKhw5DCliBdw7NWwpUsVsK/w5UhwoMJA8OIADLCgAzCoFxKP8OyTTfDmcOYZDBhABlABsKQAcOQwr0AwrfCvcOkfkJdJQEAAAAASUVORMKuQmDCgC

282 chars.

With "Base85" the length is 265.

The expected result:

iVBORw0KGgoAAAANSUhEUgAAACAAAAAgCAIAAAD8GO2jAAAAAXNSR0IArs4c6QAAAARnQU1BAACxjwv8YQUAAAAJcEhZcwAADsMAAA7DAcdvqGQAAAA+SURBVEhL7dIxCgAwCMVQ739pu/yh0JYgXfNWlSxWv9UhgwkDyAAygAygXEo/8k032dhkMGEAGUAGkAHQvQC3veR+Ql0lAQAAAABJRU5ErkJggg==

228 chars. (Math.ceil(169*4/3+2), btoa(binaryString))

With "Base85" the length should be ~212 (169*5/4).

Does it work as planned?

All chars must be distinct

Hi, I am interested in your code and I'd like to be able to consume the library as a nuget package. Do you have any official nuget package feed that can provide me with the package?

I find out something confusing between the system definition and the data in diagram.xlsx about the equation below.

Take a exmpale(as Base64):

a=64
k=4
b=2
n=24

// then, according to the equation, ratio should be
r=64^4 / 2^24 =1

However, in the Diagram.xlsx:

Am i misunderstanding something?

@KvanTTT could you explain it? 😄

Anyway, really appreciate your awesome work.

See https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/BigInt

See here for implementation: http://sourceforge.net/projects/base91/

See http://en.wikipedia.org/wiki/Ascii85 for information and samples.